CVPR 2024 机器学习方向总汇（多任务、联邦学习、迁移学习和对抗等）

1、Machine Learning(机器学习)多任务、联邦学习、迁移学习和对抗等

• Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision
  👍摘要
• Improving Physics-Augmented Continuum Neural Radiance Field-Based Geometry-Agnostic System Identification with Lagrangian Particle Optimization
  🏠project
• Circuit Design and Efficient Simulation of Quantum Inner Product and Empirical Studies of Its Effect on Near-Term Hybrid Quantum-Classic Machine Learning
• 对抗
  • Infrared Adversarial Car Stickers
  • Robust Distillation via Untargeted and Targeted Intermediate Adversarial Samples
  • Revisiting Adversarial Training Under Long-Tailed Distributions
  • PAD: Patch-Agnostic Defense against Adversarial Patch Attacks
    ⭐code
  • Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training
  • MimicDiffusion: Purifying Adversarial Perturbation via Mimicking Clean Diffusion Model对抗性扰动
  • Towards Transferable Targeted 3D Adversarial Attack in the Physical World
  • Deep-TROJ: An Inference Stage Trojan Insertion Algorithm through Efficient Weight Replacement Attack攻击
  • Attack To Defend: Exploiting Adversarial Attacks for Detecting Poisoned Models
  • Re-thinking Data Availability Attacks Against Deep Neural Networks
  • SlowFormer: Adversarial Attack on Compute and Energy Consumption of Efficient Vision Transformers
  • Re-thinking Data Availablity Attacks Against Deep Neural Networks攻击
  • NAPGuard: Towards Detecting Naturalistic Adversarial Patches
  • Focus on Hiders: Exploring Hidden Threats for Enhancing Adversarial Training
  • Not All Prompts Are Secure: A Switchable Backdoor Attack Against Pre-trained Vision Transfomers后门攻击
  • Physical Backdoor: Towards Temperature-based Backdoor Attacks in the Physical World
  • Backdoor Defense via Test-Time Detecting and Repairing
  • Nearest Is Not Dearest: Towards Practical Defense against Quantization-conditioned Backdoor Attacks
  • Semantic-Aware Multi-Label Adversarial Attacks对抗攻击
  • Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning
  • Improving Transferable Targeted Adversarial Attacks with Model Self-Enhancement对抗攻击
  • On the Robustness of Large Multimodal Models Against Image Adversarial Attacks
  • Incremental Residual Concept Bottleneck Models
  • Revisiting Adversarial Training at Scale
    ⭐code
  • Language-Driven Anchors for Zero-Shot Adversarial Robustness零样本对抗
  • Transferable Structural Sparse Adversarial Attack Via Exact Group Sparsity Training
  • Learning to Transform Dynamically for Better Adversarial Transferability
  • Defense without Forgetting: Continual Adversarial Defense with Anisotropic & Isotropic Pseudo Replay
  • Boosting Adversarial Transferability by Block Shuffle and Rotation
    ⭐code对抗性可转移性
  • MMCert: Provable Defense against Adversarial Attacks to Multi-modal Models
  • Pre-trained Model Guided Fine-Tuning for Zero-Shot Adversarial Robustness
    👍VILP
  • Adversaral Doodles: Interpretable and Human-drawable Attacks Provide Describable Insights
  • PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor
  • Revisiting Adversarial Training under Long-Tailed Distributions
    ⭐code
  • Towards Fairness-Aware Adversarial Learning
  • Dispel Darkness for Better Fusion: A Controllable Visual Enhancer based on Cross-modal Conditional Adversarial Learning
  • Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement
  • Robust Overfitting Does Matter: Test-Time Adversarial Purification With FGSM
  • Boosting Adversarial Training via Fisher-Rao Norm-based Regularization
    ⭐code
  • A Stealthy Wrongdoer: Feature-Oriented Reconstruction Attack against Split Learning攻击
  • 后门攻击
    • LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning
      ⭐code
    • Test-Time Backdoor Defense via Detecting and Repairing
    • Data Poisoning based Backdoor Attacks to Contrastive Learning
      ⭐code
• 持续学习
  • RCL: Reliable Continual Learning for Unified Failure Detection
  • Consistent Prompting for Rehearsal-Free Continual Learning
  • Improving Plasticity in Online Continual Learning via Collaborative Learning
  • Boosting Continual Learning of Vision-Language Models via Mixture-of-Experts Adapters
    ⭐code
  • Enhancing Visual Continual Learning with Language-Guided Supervision
  • Convolutional Prompting meets Language Models for Continual Learning
  • Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning
  • Orchestrate Latent Expertise: Advancing Online Continual Learning with Multi-Level Supervision and Reverse Self-Distillation
  • InfLoRA: Interference-Free Low-Rank Adaptation for Continual Learning
  • Learning Equi-angular Representations for Online Continual Learning
  • BrainWash: A Poisoning Attack to Forget in Continual Learning
  • Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning持续学习
  • Traceable Federated Continual Learning
  • Interactive Continual Learning: Fast and Slow Thinking
• 增量学习
  • Towards Efficient Replay in Federated Incremental Learning
• 类增量学习
  • Dual-Consistency Model Inversion for Non-Exemplar Class Incremental Learning
  • Class Incremental Learning with Multi-Teacher Distillation
  • Dual-Enhanced Coreset Selection with Class-wise Collaboration for Online Blurry Class Incremental Learning
  • Generative Multi-modal Models are Good Class Incremental Learners
  • FCS: Feature Calibration and Separation for Non-Exemplar Class Incremental Learning
  • OrCo: Towards Better Generalization via Orthogonality and Contrast for Few-Shot Class-Incremental Learning
  • Long-Tail Class Incremental Learning via Independent Sub-prototype Construction
  • Gradient Reweighting: Towards Imbalanced Class-Incremental Learning
  • DYSON: Dynamic Feature Space Self-Organization for Online Task-Free Class Incremental Learning
  • NICE: Neurogenesis Inspired Contextual Encoding for Replay-free Class Incremental Learning
    ⭐code
  • Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learning
    ⭐code
  • Text-Enhanced Data-free Approach for Federated Class-Incremental Learning
    ⭐code
  • Generative Multi-modal Models are Good Class-Incremental Learners
    ⭐code
  • Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning
    ⭐code
• 多任务
  • Masked AutoDecoder is Effective Multi-Task Vision Generalist
  • OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning
  • Task-conditioned adaptation of visual features in multi-task policy learning
  • DiffusionMTL: Learning Multi-Task Denoising Diffusion Model from Partially Annotated Data
    ⭐code
  • FedHCA2: Towards Hetero-Client Federated Multi-Task Learning
    ⭐code
  • MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learning
  • Joint-Task Regularization for Partially Labeled Multi-Task Learning
  • Task-Conditioned Adaptation of Visual Features in Multi-Task Policy Learning
  • 多标签学习
    • View-Category Interactive Sharing Transformer for Incomplete Multi-View Multi-Label Learning
• 多视角学习
  • Rethinking Multi-view Representation Learning via Distilled Disentangling
    ⭐code
• 元学习
  • FREE: Faster and Better Data-Free Meta-Learning
  • Improving Generalization via Meta-Learning on Hard Samples
• 联邦学习
  • An Aggregation-Free Federated Learning for Tackling Data Heterogeneity
  • Decentralized Directed Collaboration for Personalized Federated Learning
  • Rethinking the Representation in Federated Unsupervised Learning with Non-IID Data
  • Byzantine-robust Decentralized Federated Learning via Dual-domain Clustering and Trust Bootstrapping
  • FLHetBench: Benchmarking Device and State Heterogeneity in Federated Learning
  • Revamping Federated Learning Security from a Defender's Perspective: A Unified Defense with Homomorphic Encrypted Data Space
  • Unlocking the Potential of Prompt-Tuning in Bridging Generalized and Personalized Federated Learning
  • Mixed-Precision Quantization for Federated Learning on Resource-Constrained Heterogeneous Devices
  • FedSelect: Personalized Federated Learning with Customized Selection of Parameters for Fine-Tuning
  • Fair Federated Learning under Domain Skew with Local Consistency and Domain Diversity
    ⭐code
  • Global and Local Prompts Cooperation via Optimal Transport for Federated Learning
  • PerAda: Parameter-Efficient Federated Learning Personalization with Generalization Guarantees
    ⭐code
  • Relaxed Contrastive Learning for Federated Learning
  • DiPrompT: Disentangled Prompt Tuning for Multiple Latent Domain Generalization in Federated Learning
  • FedAS: Bridging Inconsistency in Personalized Federated Learning
    ⭐code
  • Leak and Learn: An Attacker's Cookbook to Train Using Leaked Data from Federated Learning
  • Data Valuation and Detections in Federated Learning
    ⭐code
  • An Upload-Efficient Scheme for Transferring Knowledge From a Server-Side Pre-trained Generator to Clients in Heterogeneous Federated Learning
    ⭐code
  • Adaptive Hyper-graph Aggregation for Modality-Agnostic Federated Learning
  • FedUV: Uniformity and Variance for Heterogeneous Federated Learning
  • FedSOL: Stabilized Orthogonal Learning with Proximal Restrictions in Federated Learning
  • Communication-Efficient Federated Learning with Accelerated Client Gradient
• 强化学习
  • Improving Unsupervised Hierarchical Representation with Reinforcement Learning
  • AlignSAM: Aligning Segment Anything Model to Open Context via Reinforcement Learning强化学习
  • Training Diffusion Models Towards Diverse Image Generation with Reinforcement Learning
  • POCE: Primal Policy Optimization with Conservative Estimation for Multi-constraint Offline Reinforcement Learning
  • DMR: Decomposed Multi-Modality Representations for Frames and Events Fusion in Visual Reinforcement Learning
  • Learning to Control Camera Exposure via Reinforcement Learning
    🏠project
  • Regularized Parameter Uncertainty for Improving Generalization in Reinforcement Learning
  • Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World
    🏠project
• 多模态机器学习
  • DIEM: Decomposition-Integration Enhancing Multimodal Insights
• 迁移学习
  • Model Inversion Robustness: Can Transfer Learning Help?
  • Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning
  • Structured Model Probing: Empowering Efficient Transfer Learning by Structured Regularization
  • UniPT: Universal Parallel Tuning for Transfer Learning with Efficient Parameter and Memory
    ⭐code
  • Initialization Matters for Adversarial Transfer Learning
• 对比学习
  • Improving Graph Contrastive Learning via Adaptive Positive Sampling
  • MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
  • BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
  • Universal Novelty Detection Through Adaptive Contrastive Learning
  • NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models
    🏠project
• 模仿学习
  • LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation
• 上下文学习
  • Skeleton-in-Context: Unified Skeleton Sequence Modeling with In-Context Learning
    ⭐code
• 弱监督学习
  • Virtual Immunohistochemistry Staining for Histological Images Assisted by Weakly-supervised Learning
• 启示学习
  • One-Shot Open Affordance Learning with Foundation Models